Nicola Shaver

A few critical things to consider during the training phase of a project: 1. Different types of training will suit different personas across the organization. Ensure that you have done the work upfront of understanding those personas and their needs, so you can tailor training accordingly. 2. If the change you're rolling out is a technology platform, leverage the vendor's own resources to support your efforts. Sometimes they will have training built into the system, other times they may have collateral you can use. 3. There are newer technology tools such as WalkMe that allow you to build training in to platforms to support your users as and when they need it. This can manifestly help with change efforts.

The support and resources required will differ by persona and the type of resistance. Learning to understand where resistance is coming from is key. For example, a leader who is used to being in control may not like the feeling of losing control when confronted with new ways of doing things. Providing coaching and training to help them regain control will ease resistance. A junior seeking promotion may resist change because they have concerns it will slow them down at a time when they need to show excellence. Providing them with reassurance that this won't affect their performance or promotional opportunities will help overcome this type of resistance. Support and resources are important - just make sure they are the right kind!

Since 2018 we have seen the rise of (and a focus on) generative models that were increasingly "large". Bigger models are not always better, though, if what you need is specialized knowledge. In the legal profession, for example, a lot of key information sits behind paywalls or on websites in formats that are not innately machine-readable. That means a large, multi-purpose model that has been trained on publicly available data won't be able to provide reliable, accurate responses to complex legal questions. This is one of the core data challenges of generative AI, but it can be overcome by shifting focus to developing targeted, specialized models for specific purposes that have been trained on the right libraries of structured, sound data.

An innovation program without a strategy won't be successful, partly because your organization's leadership won't be able to assess your team's value or success without one. In developing an effective strategy for innovation some of the key factors to consider include: - your company's overarching strategic goals, and how your innovation program can support and tie into these, - the level of relative buy-in you have from leadership for the innovation program, and whether there is interest primarily in short-term or longer term planning, - the priorities of your "end users" within the organization and beyond, and their needs, - the size and budget you have at your disposal. A strategy should be revisited and tweaked at regular intervals.

Resilience is gained through adversity, which means there's a silver lining to every downfall you experience. That means you can draw upon personal as well as professional experiences to help you learn to adapt and deal with resistance. For example, - If you have a career in which you've changed workplace several times, you will have learned to develop new relationships and adapt to new process and work environment. - If you've moved countries, you will have learned to adapt to entirely cultures and even, perhaps, new languages. Adjusting change techniques to serve the particular complexities of a specific project is not so different. Understand where the resistance comes from, what concerns drive it, and adapt techniques accordingly.

In addition to pinning down specific objectives and metrics for testing, it's important to set a cadence of testing that comports with the "share early and often" adage. Hearing from users only once during development, for example, won't ensure you get a result that meets user needs. Sharing and gaining feedback throughout the development process is critical.